Access to eye care during the COVID-19 pandemic, India

Abstract Objective To study the impact of the coronavirus disease 2019 (COVID-19) pandemic on outpatient visits to eye care facilities in south India. Methods We used data on 7.69 million outpatient visits to primary (i.e. vision centres), secondary and tertiary Aravind Eye Care System’s centres between January 2019 and June 2021. We compared outpatient numbers and outpatients’ age and sex between the pandemic period and the pre-pandemic period in 2019 for all centres, whereas vision and ophthalmic assessments were compared for vision centres only. Findings During the first wave, the number of outpatient visits at tertiary, secondary and vison centres was 39% (647 968/1 656 296), 60% (170 934/283 176) and 73% (180 502/246 282) respectively, of 2019 levels. During the second wave, outpatient visits at tertiary, secondary and vision centres were 54% (385 092/710 949), 73% (88 383/121 739) and 79% (121 993/154 007), respectively, of 2019 levels. The proportion of outpatients who were female or younger than 20 years or older than 60 years was significantly lower during the first and second waves than in 2019 (P < 0.0001 for all). The proportion of outpatients whose worse eye vision was poorer than 5/60 or who required referral was significantly higher (P < 0.0001 for both). Conclusion Restrictive measurements led to declines in outpatient visits, however the decline was less at secondary and vision centres than at tertiary centres. Easy access to specialized ophthalmic care via telemedicine and the relative proximity of these centres to communities helped reduce barriers to access.


Introduction
Although the coronavirus disease 2019 (COVID-19) pandemic generated tremendous suffering worldwide, it also provided an opportunity to study patient behaviour. During the early stages of the pandemic, many countries adopted stringent measures to contain the disease. In India, emergency health-care services continued but regular outpatient services were suspended and elective surgery deferred. Despite the stepwise reopening of outpatient eye care following expert committee guidelines, 1 we witnessed a drastic decline in clinic visits and procedures, which reflected travel restrictions, unemployment-related financial challenges and fear of infection. During the acute phase of the pandemic, patient numbers at one tertiary eye care hospital in India fell to a mere 3.5% of the previous year's figure. 2,3 A similar pattern was seen in the United States of America (USA) and was possibly repeated worldwide. 4 Although the pandemic adversely affected access to eye care, the number of sight-threatening conditions occurring would nevertheless have been expected to remain the same and any delays in care could have led to an increasing number of individuals becoming irreversibly blind or experiencing a reduced quality of life.
In India, hospitals were able to remain open for emergencies and critical care. Secondary and tertiary hospitals belonging to the Aravind Eye Care System network in south India, which handle 4.6 million outpatients per year, were among the few eye hospitals in the country that remained open throughout the pandemic. Ophthalmic care was also provided by private multispecialty hospitals but they were overwhelmed by the inflow of COVID-19 patients. Moreover, most private eye clinics were closed during the acute phases of the pandemic. 2 Outreach eye camps run by the Aravind Eye Care System, which cater to rural populations, also ceased operating due to the restrictive measurements. 2,5 In contrast, the Aravind Eye Care System's primary eye-care centres, known as vision centres, started to function in a phased manner in accordance with local restrictions. With travel restrictions in force to contain the spread of COVID-19, it seemed logical that patients would prefer to seek care locally.
The aim of this study was to examine the overall impact of travel restrictions, the closure of eye-care facilities, evolving patient preferences and other challenges associated with COVID-19 on the volume and nature of outpatient visits to primary, secondary and tertiary levels of eye care in south India. We hypothesized that patients may have accessed vision centres more often in these unprecedented times by virtue of their proximity to communities. Our analysis compared outpatient attendance at all facilities in the Aravind Eye Care System and the severity of presenting eye conditions at vision centres during the COVID-19 pandemic with data for the preceding year to understand changes in the way patients accessed care.

Methods
We conducted a cross-sectional study of 87 eye-care centres belonging to the Aravind Eye Care System in the states of Tamil Nadu and Pondicherry in south India: (i) six were tertiary eye care hospitals equipped to cater for the entire spectrum of eye-care services; (ii) six were secondary eye-care hospitals Objective To study the impact of the coronavirus disease 2019 (COVID-19) pandemic on outpatient visits to eye care facilities in south India. Methods We used data on 7.69 million outpatient visits to primary (i.e. vision centres), secondary and tertiary Aravind Eye Care System's centres between January 2019 and June 2021. We compared outpatient numbers and outpatients' age and sex between the pandemic period and the pre-pandemic period in 2019 for all centres, whereas vision and ophthalmic assessments were compared for vision centres only. Findings During the first wave, the number of outpatient visits at tertiary, secondary and vison centres was 39% (647 968/1 656 296), 60% (170 934/283 176) and 73% (180 502/246 282) respectively, of 2019 levels. During the second wave, outpatient visits at tertiary, secondary and vision centres were 54% (385 092/710 949), 73% (88 383/121 739) and 79% (121 993/154 007), respectively, of 2019 levels. The proportion of outpatients who were female or younger than 20 years or older than 60 years was significantly lower during the first and second waves than in 2019 (P < 0.0001 for all). The proportion of outpatients whose worse eye vision was poorer than 5/60 or who required referral was significantly higher (P < 0.0001 for both). Conclusion Restrictive measurements led to declines in outpatient visits, however the decline was less at secondary and vision centres than at tertiary centres. Easy access to specialized ophthalmic care via telemedicine and the relative proximity of these centres to communities helped reduce barriers to access. included strict adherence to mask-wearing, social distancing and hand hygiene. 6 We reviewed the electronic medical records of all consecutive outpatients seen at the facilities between January 2020 and June 2021. To study trends in outpatient behaviour and characteristics, we considered four time periods: (i) the period immediately before the COVID-19 pandemic from 1 January to 23 March 2020 (i.e. the pre-COVID-19 period); (ii) the first wave of the pandemic from 24 March to 31 October 2020; (iii) the period after the first wave from 1 November 2020 to 31 March 2021; and (iv) the second wave from 1 April to 30 June 2021. Table 1 lists the COVID-19 restrictions in place at study sites during different phases of the pandemic. Data for these time periods were compared with data for corresponding periods in 2019, matched by day of the week and month (Table 1). Comparative 2019 data for each vision centre included only data for dates corresponding to those dates in 2020 and 2021 when that centre was open. Vision centres established in 2020 or 2021 were not included in the study. The patients' demographic characteristics, including age and sex, across all facilities were analysed specifically for the first and second waves of the pandemic. In addition, presenting vision assessments and clinical diagnoses were studied only in vision centres and only during the two waves of the pandemic. Although these two waves spanned 15 months across 2020 and 2021, the comparison period covered only the corresponding days in 2019, the year immediately preceding the COVID-19 pandemic (Table 1).
Variables extracted from medical records for each visit included the date of the visit and the patient's age and sex. For patients attending vision centres, additional information was obtained on visual acuity at presentation and on the diagnosis for the worse eye during the first and second waves and the respective comparison periods. If there was more than one diagnosis, the diagnosis that most threatened vision was considered the primary diagnosis. For example, if a patient had an immature cataract and retinal detachment, the primary diagnosis was retinal detachment.
To understand changes in patient behaviour during the pandemic, we contacted a random subset of patients by phone and, after obtaining informed consent orally, asked them why they delayed or missed consultations. Similarly, we spoke to patients who attended hospitals to determine what motivated their visit during the COVID-19 pandemic.
We followed the tenets of the Declaration of Helsinki and obtained ethical clearance from the institutional ethics committee of the Aravind Eye Hospital in Pondicherry.

Statistical analysis
Data were saved on Excel (Microsoft Corporation, Redmond, USA) and analysed using Stata v.14.0 (StataCorp LLC, College Station, USA). We compared differences in variables between the pandemic and pre-pandemic periods using two-sample proportion tests. A P value less than 0.05 was considered statistically significant.

Results
We obtained data on 7.69 million outpatient visits to 87 Aravind Eye Care System centres between 1 January 2019 and 30 June 2021 (  Fig. 1 shows a moving average of the number of outpatient visits between March 2020 and June 2021 as a proportion of the number during the comparison period for each eye care facility level. During phase I in the first wave of the pandemic from 24 March 2020 to 3 May 2020, when there was no transport (Table 1), the outpatient volume at tertiary and secondary hospitals was only 5.3% (16 019/300 590) and 9.5% (4825/50 656) of 2019 volumes, respectively; vision centres were closed. During phases II, III and IV in the first wave, outpatient volumes gradually increased across all levels of care. In general, the volume increased more rapidly at vision centres than at tertiary or secondary centres, except in phase IV, when the volume at vision centres during the study period as a proportion of that during the 2019 comparison period fell to 58% (18 722/32 321) overall because several vision centres were located in newly declared containment zones where movement was restricted. Between 1 September and 31 October 2020, when all travel restrictions were lifted (Table 1), outpatient volumes increased steadily across all levels of care. During the second wave, from 10 May to 30 June 2021, vision centres again performed better than tertiary or secondary care centres. In addition, when all travel restrictions had been lifted after phase V, tertiary centres had only 66% (150 540/229 013) of the previous year's volume on average, whereas vision centres and secondary centres recovered to 88% (47 957/54 714) and 86% (32 833/37 981), respectively. After the first wave, a rapid bounce back was seen across all levels of eye care. This was followed by another drastic decline in patient volume during the restrictions in the second wave, although vision centres performed better. Table 3 shows the age and sex distributions, respectively, of outpatients visiting vision centres and secondary and tertiary care centres during the first and second waves of the COVID-19 pandemic and during the comparison periods. Though the absolute number of visits was lower during the first and sec-  (Table 1). b For the comparison period, dates in the study period were matched with dates in 2019, the last year before the pandemic (Table 1). c The 1 674 364 outpatient visits made between January and June 2019 were used twice for comparisons. Consequently, this figure was deducted when the total number of unique outpatient visits was calculated. d P values were derived using the two-sample proportion test. e Vision centres are primary care centres.
ond waves than during 2019 overall, the proportion of all patients who were aged 21-60 years was significantly higher during the pandemic across all facilities (P < 0.0001 for all). Correspondingly, the proportion of all patients who were aged 61 years and older or 20 years and younger was lower during the pandemic across all levels of eye care (P < 0.001 for all). In addition, the proportion of patients who were female was significantly lower during the pandemic compared to 2019 across all facilities (P < 0.0001 for all). Table 4 and Table 5 show the distributions of visual acuity and ophthalmic diagnosis at presentation, respectively, recorded on outpatient visits to vision centres during the first and second waves of the pandemic and during the comparison periods in 2019. The proportion of patients who presented with a visual acuity of 5/60 to 3/60 or of less than 3/60 in the worse eye was significantly higher during the pandemic period than during 2019 (P < 0.0001 for both), though the absolute number of visits was smaller (Table 4). In addition, there was a substantial shift in the distribution of diagnoses between 2019 and the pandemic period ( Table 5). The proportion of diagnoses that required a referral to a tertiary care hospital (e.g. lens-induced glaucoma, cataract, corneal infection, ocular trauma, uveal disease and retinal disease such as diabetic retinopathy and vein occlusion) was significantly higher during the pandemic (P < 0.0001 for all). In contrast, the proportion of diagnoses of glaucoma (other than lens-induced glaucoma) or refractive error was significantly lower (P < 0.0001 for both).
In the phone survey of a random subset of patients who were asked why they did not attend their regular followup, 70% (329/470) mentioned travel restrictions, 35% (164/470) feared getting infected, 13% (61/470) had financial problems, 8.9% (42/470) did not have an escort, 8.7% (41/470) did not feel it was necessary and 17% (79/470) had a consultation elsewhere. These responses were not mutually exclusive. Individuals who attended hospital during the active phase of the pandemic said they sought care because they either were afraid of going blind, had worsening symptoms or anticipated future travel restrictions.

Discussion
The COVID-19 pandemic provided us with a unique opportunity to study how ease of access affects outpatient visits to primary, secondary and tertiary eye care facilities. Our study, which involved a large number of patients, found that vision centres experienced a smaller decline in outpatient numbers during both waves of the pandemic than secondary or tertiary eye hospitals. These findings reinforce the importance of ease of access for the utilization of care, particularly in underserved and rural communities.
The significant decline in patient numbers we observed during the pan-  (Table 1). Travel restrictions during the first wave of the pandemic were applied in five phases (for details, see Table 1). Vision centres were closed during phase I. Although the second wave started on 1 April 2021, travel restrictions were not imposed until 10 May 2021.
demic was very different from normal fluctuations. In previous years, the patient volume grew consistently and gradually at an annual rate of 2 to 5% across all levels of eye care, with higher growth rates in newly established hospitals and vision centres. Nearly 40% of patients presenting to the Aravind Eye Care System's tertiary hospitals travelled over 100 km, largely by public bus, with an average travel time of 4 hours. 7 In contrast, travel to secondary hospitals and vision centres generally took less time, which made them easier to access, especially during the pandemic. Additionally, these facilities tended to have shorter turnaround times and the realtime teleconsultations available at vision centres eliminated the potential risk of exposure to infection during travel.
Our results align with those of previous studies into the effect of the COVID-19 pandemic on patient care across all health-care facilities. [8][9][10][11][12][13][14] During phase I in the first wave, when all vision centres were closed, outpatient volumes in tertiary and secondary centres were only 5.3% and 9.5% of corresponding volumes in 2019, respectively. The decline occurred despite our efforts to make the public aware through social media and newspapers that hospitals continued to function with the necessary safety precautions in place. During both waves of the pandemic, staff were assigned to reminding patients about pending care and especially about urgent or emergent care. We also helped patients get the mandatory passes needed to travel to hospital.
When we investigated the age profile of outpatients, we observed that the proportion who were dependent on others for their care (i.e. those aged under 20 years and those aged over 60 years) was lower during the pandemic period than in 2019. A previous study has shown that the proportion of children and adolescents visiting ophthalmology emergency departments decreased from 10 to 5.3% during the early phases of the COVID-19 pandemic. 13 Possible reasons for this trend include: (i) fear of infection among adolescents and elderly people; (ii) patients needing someone to escort them; and (iii) comorbid conditions and decreased mobility among the elderly. 14 Our study also demonstrated that the proportion of female patients decreased significantly across all eyecare levels during the pandemic relative  15 A sex bias in the utilization of health care in general had been observed in China and India before the COVID-19 pandemic. [15][16][17] Telemedicine became increasingly important during the pandemic for both health care as a whole and ophthalmology. [18][19][20] At our vision centres, routine tele-ophthalmic consultation was advantageous for both health-care providers and patients: access to care improved and the need for travel, with its associated costs and risk of exposure to the virus, decreased. In recognition of the importance of ease of access, we are now systematically shifting care to the local level (e.g. to vision centres), where appropriate. For example, over 55% of patients who have undergone cataract surgery are now seen at the nearest vision centre for postoperative follow-up on day 1. The decentralization of care has also reduced the carbon footprint associated with travel. In addition, we are increasing the scope of telemedicine at our vision centres by integrating the telemedicine process into electronic medical records and facilitating consultations in some subspecialities, which will make care at vision centres more comprehensive. We believe that the processes triggered by COVID-19 will bring lasting improvements. The effective utilization of telemedicine in primary care has already produced highly encouraging results. Moreover, patients' real-time teleconsultations with ophthalmologists provide opportunities for continuing clinical education for vision technicians at vision centres, which can reinforce their skill development. Still, continuing training of staff at secondary and primary care levels is important for improving the quality of diagnosis and referrals. This model of augmenting primary eye-care centres with telemedicine has the potential to revolutionize the entire health-care system.
Our study has several strengths: (i) it involved a large data set of 7.69 million outpatient visits across all levels of eye care; (ii) it covered both the first and second waves of the COVID-19 pandemic; and (iii) it compared pandemic and pre-pandemic periods. Additionally, the Aravind Eye Care System does not have an appointments system, which eliminates provider-related biases and, thus, ensures that data reflect the natural change in patient numbers. Our study was limited by its retrospective nature and by a lack of data on presenting morbidity patterns at tertiary and secondary eye-care hospitals. Moreover, we did not analyse compliance with referrals from vision centres to secondary and tertiary hospitals. Finally, our study was done in south India, which could limit the generalizability of our results because pandemic restrictions varied widely across different parts of the world.
In conclusion, we believe the trend we observed of patients seeking care for both emergency and non-emergency conditions closer to home will be relevant after the pandemic, particularly now that specialist consultations are  Table 3 because visual acuity was not documented in some case records.  Table 3 because the diagnosis was not documented in some case records.